Over the past two decades, the in vitro amplification of specific nucleic acids has become an essential tool for molecular biologists. More recently, multiplexed amplification, in which a plurality of nucleic acid sequences are amplified in a single reaction, Chamberlain et al., Nucl. Acid Research 16(23):11141-1156 (1988); U.S. Pat. No. 5,582,989, has become increasingly important. For example, multiplexed amplification, particularly multiplexed polymerase chain reaction (PCR), has been used to provide genetic fingerprints of infectious disease organisms. Other applications, such as multiplex SNP genotyping and variation scanning (for example, by mismatch repair detection), also greatly benefit from PCR multiplexing.
In its original implementation, multiplex PCR reactions include a specific primer pair for each locus to be amplified. These approaches have been plagued with problems, however, including uneven or failed amplification of some templates (especially those having GC rich-sequences), preferential amplification of other templates, poor sensitivity and specificity, poor reproducibility, and the generation of spurious amplification products (Henegariu et al., BioTechniques 23(3): 504-511 (1997); Markoulatos et al., J. Clin. Lab. Anal. 16: 47-51 (2002)).
Various modifications to the original approach have been developed in efforts to minimize these problems. Among these modifications are changes to the reaction conditions, including adjustment of primer concentrations, MgCl2 and dNTP concentrations, changes in PCR buffer concentrations, balance between MgCl2 and dNTP concentrations, amounts of template DNA and Taq DNA polymerase, extension and annealing time and temperature, and the addition of adjuvants (Henegariu et al., BioTechniques 23(3): 504-511 (1997); Markoulatos et al., J. Clin. Lab. Anal. 16: 47-51 (2002)). Other strategies used include subcycling temperatures between high and low temperatures below the denaturation temperature, used during the annealing and elongation steps (U.S. Pat. No. 6,355,422), and the use of one sequence-specific primer and one common primer (Broude et al., Proc. Natl. Acad. Sci. USA 98, 206-211 (2001))
The intractability of GC-rich sequences to multiplex PCR has also been addressed by a method in which addition of betaine and dimethylsulfoxide (DMSO) to the PCR reaction mix is said to allow more uniform amplification from a heterogeneous population of DNA molecules, many of which were GC-rich (Baskaran et al., Genome Research 6: 633-638 (1996)).
Yet other approaches alter the primers. In one such effort, chimeric oligonucleotides are used as primers: the oligonucleotides include a 3′ domain that is complementary to template, conferring template specificity, and a 5′ domain that is noncomplementary to template; the 5′ domain includes a sequence used to prime extension in rounds of PCR amplification subsequent to the first. In this latter scheme, however, the cycles of amplification following the first amplify whatever product is generated in the first cycle, whether correct or erroneous. Thus, while the technique allows for more uniform amplification, it does not address the problem of spurious products.
In an analogous approach designed to clone the shared components in two complex samples, Brookes et al., Human Molec. Genetics 3(11):2011-2017 (1994), ligate primers to template ends generated by restriction fragment digestion. None of the above-mentioned approaches, however, fully solves the problems associated with multiplex PCR. Thus, there is a continuing need in the art for a method that allows the specific and uniform amplification of multiple nucleic acid sequences in a single reaction, without the generation of spurious products.
Multiplex targeted genome amplification allows simultaneous generation of many targets in the same tube for cost-effective genotyping, sequencing or resequencing. The most powerful targeted amplification has been the polymerase chain reaction (PCR). Traditional multiplex PCR has been used to amplify two or more targets by putting multiple pairs of primers simultaneously in the same reaction. However, due to exponential increase in primer-dimer interaction when more pairs of primers are included as well as unequal amplification rates among different amplicons, the multiplexing level of this traditional scheme is typically efficient for up to about 20-plex, often with individual primer concentrations requiring adjustment. This scheme has found applications in multiplex real time PCR or microsatelite amplification and commercial kits, for example, from Qiagen, are available.
Current amplification methods range from non-specific amplification of the entire genome, for example, whole genome amplification (WGA) methods such as MDA, to highly targeted PCR amplification of a few or a single selected region of, for example, a few kb. Methods that result in amplification of a reproducible subset of a genome, for example, the Affymetrix whole genome sampling assay (WGSA) may also be used to amplify genomic material for downstream analysis. The WGA methods generally result in a non-selective amplification of the entire genome. The WGSA method results in amplification of a selected subset of the genome, the subset being defined by the restriction enzyme or enzymes used for cutting the DNA prior to adapter-mediated PCR amplification. Other methods that allow targeted amplification of large numbers of specific targets include, for example, the With whole genome amplification methods being applied to amplifying the whole human genome (a few billion bp) at one end and PCR in targeting a few kb sequences in the other end, there is a need to have a strategy amplifying 1-100 million by that can cover exons and promoter regions of most or all the functional genes.
Attempts have been made over the years since the invention of PCR to increase the multiplex level of PCR. Some of the strategies include two-stage PCR with universal tails (Lin Z et al., PNAS 93: 2582-2587, 1996; Brownie J. et al., Nucleic Acids Res. 25: 3235-3241, 1997), solid-phase multiplex PCR (e.g., Adams and Kron, U.S. Pat. No. 5,641,658; Shapero et al., Genome Res. 11: 1926-1934, 2001), multiplexed anchored runoff amplification (MARA, Shapero et al., Nucleic Acid Res. 32: e181, 2004 and U.S. Pat. No. 7,108,976), PCR with primers designed by a special bioinformatical tool (Wang et al., Genome Res. 15: 276, 2005), selector-guided multiplex amplification (Dahl F et al., Nucleic Acids Res. 33: e71, 2005), and dU probe-based multiplex PCR after common oligo addition (Faham M and Zheng J, U.S. Pat. No. 7,208,295 and Faham M et al., PNAS 102: 14717-14722, 2005). However, most of above strategies are either work most efficiently at about 100 to 1000-plex, or suffer low efficiency, with the exception of the last two strategies that are potentially scalable to over 10,000-plex (or over a million bp). The method of Dahl et al. requires synthesis of long oligo probes (usually >80 bases) and the method of Faham et al. requires synthesis of dU probes by PCR for each target (Faham M et al., 2005). Multiplex PCR methods are also disclosed in U.S. Patent publication Nos. 20030104459. See also, Nilsson et al., Trends. Biotechnol. 24(2):83-8, 2006 and Stenberg et al., NAR 33(8):e72, 2005.